Transcriptional and mutational analysis of the Helicobacter pylori urease promoter

Abstract

Urease is an essential virulence factor of the human gastric pathogen Helicobacter pylori, and is expressed to very high levels. The promoter of the urease operon contains sequences resembling the canonical −10 and extended −10 motifs, but no discernible −35 motif. To establish the role of different motifs and regions in the urease promoter, we fused the urease promoter to a genomic lacZ reporter gene in H. pylori, made substitutions in the aforementioned promoter motifs, and also made deletions in the upstream sequences removing regulatory sequences. Substitutions in the −10, extended −10 and predicted −35 motifs all significantly altered expression of the lacZ reporter gene, demonstrating their importance in transcription of the H. pylori urease operon. In contrast, sequential deletions upstream of the −35 region did not affect expression of the lacZ reporter gene. This demonstrates the modular structure of the H. pylori urease promoter, where basal levels of transcription are initiated from a typical σ⁷⁰ promoter, which requires −10 and extended −10 motifs, and also its −35 motif for efficient transcription. Upstream sequences are not involved in basal levels of urease transcription, but play an important role in responses to environmental stimuli like nickel.

Helicobacter, Urease promoter, Palindrome, Reporter gene, Mutational analysis

1 Introduction

The Gram-negative pathogen Helicobacter pylori is now well established as a causative factor in the development of chronic gastritis and peptic ulcer disease of the human stomach, with long-term infection strongly associated with the development of gastric adenocarcinoma[1]. One of the major pathogenicity factors of H. pylori is the urease enzyme, which hydrolyses urea resulting in the formation of ammonia and carbon dioxide. Urease enzyme activity contributes to acid resistance, epithelial cell damage, chemotactic behaviour, and nitrogen metabolism[2]. The urease enzyme is essential for H. pylori infection, as urease-negative mutants are unable to colonise the gastric environment[3].

The urease genes of H. pylori are arranged in two adjacent operons: the upstream operon containing the structural urease genes, ureAB, and the second containing the accessory genes, ureIEFGH[2,4]. The UreEFGH proteins are required for activation of urease by incorporating nickel ions into the urease apoenzyme, while UreI functions as an acid-dependent urea channel[5]. The urease enzyme of H. pylori is expressed at very high levels, comprising up to 10% of the total cell protein[6], and this, along with the possibility of urease activity becoming detrimental to the bacteria at high pH[7], is a plausible reason for the expression of such a potential burden to be regulated. To date the urease genes of H. pylori have been shown to be regulated by differential mRNA decay in response to changing pH and by varying the availability of the urease cofactor nickel [8,,10].

Initiation of transcription has been extensively studied in Escherichia coli and Bacillus subtilis, where the housekeeping sigma factor σ⁷⁰ recognises specific promoter sequences located at positions −10 and −35 relative to the transcription start point (TSP). A predictive analysis of the H. pylori genome proposed that H. pyloriσ⁷⁰ shares the consensus recognition sequence with E. coliσ⁷⁰ at the −10 region[11]. A second putative promoter motif (TTAAGC), situated 19–23 bp further upstream of the −10 region in the majority of promoters, was also identified and appears unique to H. pylori[11,12].

Analysis of the sequence upstream of the H. pylori ureA gene indicated the presence of a putative σ⁵⁴-dependent promoter located 293 bp upstream of the ureA start codon[4]; however, the TSP of the urease operon was independently mapped at 54/55 bp and 56/57 bp upstream of the ureA gene [13,14]. This allowed the identification of a putative −10 sequence (TACAAT), but no obvious −35 consensus sequence [13,14].

A better understanding of the function of the different regions in the urease promoter is clearly required in order to understand the initiation of transcription and regulation of gene expression in H. pylori. Therefore, the aim of this study was to investigate the role of the different promoter motifs of the H. pylori urease promoter in transcription of the urease genes.

2 Materials and methods

2.1 Bacterial strains, media, and growth conditions

Bacterial strains and plasmids used in this study are listed in Table 1. E. coli strains were cultured at 37°C in Luria–Bertani medium with addition of kanamycin (50 μg ml⁻¹) or chloramphenicol (20 μg ml⁻¹) when required. H. pylori strains 1061 and NCTC 11637 (National Collection of Type Cultures) were routinely maintained on brain heart infusion agar (Oxoid), supplemented with vancomycin (10 μg ml⁻¹) and 2% (v/v) newborn calf serum (NCS) (Gibco BRL). Liquid cultures were grown in Brucella broth supplemented with 3% NCS. H. pylori cultures were incubated in a variable atmosphere incubator (Don Whitley Scientific) in 5% O₂, 8% CO₂ and 87% N₂ at 37°C. When appropriate, antibiotic selection in H. pylori was carried out by supplementing media with 25 μg ml⁻¹ kanamycin.

1

H. pylori strains and plasmids used in this study

	Genotype or relevant characteristics	Source/reference
H. pylori
1061	wild-type strain	[24]
BJD3.3	1061 ureA::lacZ KmR	[9]
BJD3.4a	1061 ureA(Δ+88/−112)::lacZ KmR	This study
BJD3.5b	1061 ureA(−10 motif: TACAAT→TATAAT)::lacZ KmR	This study
BJD3.6b	1061 ureA(extended −10 motif: TG→TT)::lacZ KmR	This study
BJD3.7b	1061 ureA(−35 motif: TTAATC→ACGCGT)::lacZ KmR	This study
BJD3.8a	1061 ureA(Δ−50/−112)::lacZ KmR	[10]
BJD3.9a	1061 ureA(Δ−50/−90)::lacZ KmR	[10]
BJD3.10a	1061 ureA(Δ−50/−70)::lacZ KmR	[10]
E. coli
ER1793	host strain for pBW derived constructs	New England Biolabs
Plasmid
pBW	H. pylori promoter-probe vector, KmR	[19]

^aureA(Δ+88/−112): ureA promoter lacking the sequences from positions −112 to +88, relative to the C in the experimentally determined TSP.

^bureA(−10 motif: TACAAT→TATAAT): promoter motif with the underlined residue(s) substituted.

2.2 Recombinant DNA and RNA techniques

DNA manipulations were carried out using standard techniques[15]. Plasmid DNA was isolated using a Qiaprep miniprep spin column kit (Qiagen) and H. pylori chromosomal DNA was isolated using standard methods[15]. PCR reactions were carried out using Vent DNA polymerase (New England Biolabs) following manufacturer's instructions.

Primer extension using total H. pylori RNA and primer 5′-TTGGAGTGATAATGGTGGCC-3′ was carried out to identify the urease transcription start site as previously described[16]. Yeast tRNA (Sigma) served as a negative control. Nucleotide sequencing reactions were carried out in parallel using the Thermo Sequenase radiolabelled terminator cycle sequencing kit (Amersham Biosciences), and a purified PCR product as template. The experiment was repeated on at least three independent occasions.

2.3 Construction of H. pylori wild-type and mutant ureA::lacZ derivatives

Inverse PCR mutagenesis[17] and site-directed mutagenesis by the gapped duplex method[18] were used to mutate the H. pylori 1061 ureA promoter region as cloned in pBJD3.3[9]. Correct replacement or removal of the wild-type sequences was confirmed by DNA sequencing. The mutated promoter constructs were subsequently cloned as Bam HI–Bgl II fragments in the unique Bgl II site in front of the promoterless lacZ gene of plasmid pBW[19], resulting in plasmids pBJD3.3 to pBJD3.10. These plasmids were transformed into H. pylori strain 1061 by natural transformation[20], and transformants selected for kanamycin resistance[19], resulting in H. pylori strains BJD3.3 through BJD3.10 (Table 1). Kanamycin-resistant transformants contain the pBJD derivative inserted into the H. pylori 1061 chromosome via single homologous recombination [9,19]. The resulting strain contains two copies of the ureA promoter: the mutated version preceding the promoterless lacZ gene, and the wild-type promoter preceding the intact urease operon[9].

Correct insertion of the plasmid into the H. pylori chromosome was confirmed by PCR and Southern hybridisation using the DIG-DNA Labeling and Detection Kit (Roche). Sequence analysis of the chromosomal PCR product was performed to confirm that the single cross-over event needed to integrate the vectors had occurred upstream of the mutated promoter regions and had resulted in transcriptional fusions between the mutated promoter regions and the lacZ gene. β-Galactosidase expression was measured from broth cultures as described previously, and expressed in Miller units [9,15]. Statistical analysis was performed using Student's t-test for matched pairs.

3 Results and discussion

3.1 Transcription of the urease genes is from one TSP

The ureA gene is preceded by a promoter region which contains a putative σ⁵⁴ recognition sequence and a valB tRNA sequence[4]. Since two different TSPs had already been reported [13,14], we used primer extension starting within the ureA gene and also starting from the predicted promoter region, to elucidate the transcriptional structure of the urease promoter in H. pylori strain NCTC 11637. Using the ureA internal primer, the transcription was initiated 55/56 bp upstream (TCCAAC) of the ureA start codon (Fig. 1, left hand panel), close to the positions reported previously [13,14].

1

Mapping the transcription start site of the urease operon (left hand panel) and the valB tRNA (right hand panel) positioned upstream of ureA. The non-coding strand, ordered A, C, G, T, is shown. Lanes 1, primer extension using the tRNA control; lanes 2, primer extension using H. pylori RNA. The arrow(s) on the right hand side of each panel marks the primer extension products representing the transcription start points.

To determine whether any alternative TSP was involved in urease expression, primer extension analysis was carried out further upstream of ureA, spanning the region from the lspA stop codon to the ureA ATG start codon. Only the TSP of the valB tRNA sequence[4] positioned between 252 and 325 bp upstream of the ureA start codon was identified (Fig. 1, right hand panel). Together with the fact that urease expression and activity is not altered in a H. pyloriσ⁵⁴ (rpoN) mutant (N. Dorrell, unpublished results), our data confirm that the previously proposed σ⁵⁴-dependent promoter-like sequence is not functional, and thus we conclude that urease transcription is independent of the H. pyloriσ⁵⁴ RNA polymerase complex.

3.2 Sequences upstream of the −35 motif are not required for urease transcription

The sequence upstream of the urease −35 region contains an imperfect palindromic sequence TTAAATAAT-A-ATTAGTTAA[14], which was recently demonstrated to be essential for nickel-responsive induction of urease transcription [9,10]. To assess the role of this upstream region in urease transcription, we constructed 60-, 40- and 20-bp deletions in this upstream region, thus removing the palindrome and associated sequences (Table 1, Fig. 2A). None of the deletions significantly affected activity of the β-galactosidase reporter (Fig. 2B), indicating that the upstream region is not involved in transcription from the urease promoter under standard laboratory conditions. Since the upstream sequences are involved in induction of transcription above this basal level[10], this strongly suggests that the urease promoter has a modular structure: the sequence from +1 to approximately −50 which is the standard promoter, and the upstream region including the palindrome which is responsible for further activation of the urease promoter upon certain environmental stimuli including increased nickel bioavailability[10].

2

Mutational analysis of the urease promoter region. A: Schematic representation of the mutant constructs. BJD3.3 represents the wild-type promoter, with the −10, extended −10 and −35 motif sequences given. The convergent black arrows represent the putative palindrome upstream of the −35 region. valB is the valB tRNA sequence, ureA′ and lacZ represent the ureA::lacZ transcriptional fusion. BJD3.4 through 3.10 represent the different alterations in the urease promoter region; only the changes to BJD3.3 are indicated. Asterisks indicate nucleotide substitutions, the changed nucleotide is also underlined. B: Effect of urease promoter alterations measured by β-galactosidase activity. Error bars denote S.D.; data shown are derived from four independent experiments. An asterisk denotes a significant change in β-galactosidase activity of a promoter mutation when compared to the wild-type urease promoter (P<0.01, Student's t-test for paired samples). 1061 is the H. pylori strain used for the experiments and is included as negative control.

3.3 Experimental validation of H. pylori urease promoter motifs

The urease promoter contains a typical −10 promoter motif (TACAAT), at an appropriate distance upstream from the transcriptional start site. No obvious canonical −35 consensus sequence was observed, but an extended −10 TG motif is located 1 bp upstream of the −10 sequence. The importance of these putative urease promoter motifs was experimentally investigated by generating substitutions and deletions in these regions (Fig. 2A), and subsequent transcriptional fusions of the mutated promoters to a promoterless lacZ reporter gene [9,19]. The resulting transcriptional fusions were integrated into the genome of H. pylori strain 1061 by single homologous recombination [9,19]. Effects of mutations made in the promoter region could then be quantitatively assessed based on the activity of the β-galactosidase enzyme produced, while transcription and activity of the urease operon was not affected[9].

The H. pylori urease promoter contains a suboptimal −10 region (TACAAT) when compared to the consensus E. coli sequence (TATAAT). Optimisation of the urease −10 motif to the E. coli consensus in strain BJD3.5 significantly increased β-galactosidase activity approximately 2.5-fold (Fig. 2B). Thus, even though the urease structural subunits are highly expressed in H. pylori, the −10 promoter binding region is not optimised for maximum gene transcription, as alteration of the TACAAT sequence to TATAAT leads to even higher levels of transcription. No explanation for this sub-optimal configuration of the promoter is apparent, but it suggests that there may be a penalty for the organism associated with unregulated overexpression of the enzyme.

The importance of extended −10 TG motifs in transcription initiation of both highly expressed genes and promoter sequences with weak or absent −35 consensus sequences has been well documented in other bacteria [21,22]. This is confirmed by the effect of mutating the extended −10 TG motif to a TT sequence in strain BJD3.6, which abolished β-galactosidase activity almost completely (Fig. 2B). The extended −10 TG motifs seem to play an important role in H. pylori transcription, since more than a third of experimentally validated H. pylori promoters have this motif [12,23]. Here we have now confirmed that such a extended −10 TG motif is also essential for transcription of the urease operon. It is striking that even without the TG motif, the −10 region is near-consensus yet has low activity, suggesting that other features of the overall promoter region might also modulate expression from the −10 sequence per se.

Finally, the predicted −35 motif (TTAATC) does not match the E. coli consensus sequence, but is similar to the predicted canonical H. pylori−35 motif[11]. Alteration of all bases of the urease −35 motif to ACGCGT in strain BJD3.7 completely abolished activity of the β-galactosidase reporter (Fig. 2B), thus confirming its importance in urease transcription. Single-basepair mutations will be required to further establish the exact role of the −35 motif sequences on urease transcription.

In summary, we have demonstrated that the urease promoter contains recognition sequences which are very similar to the consensus sequences of the E. coliσ⁷⁰ promoter binding motifs at the −10 and extended −10 positions. The H. pylori−35 region, whilst not having sequence homology to the E. coli−35 consensus sequence, remains important in urease transcription initiation. The transcriptional fusions created for this study may help to further characterise the molecular responses required for H. pylori to survive in its hostile niche in the gastric mucosa.

Acknowledgements

We thank Jon L. Hobman and Stephen P. Kidd for their advice on the gapped duplex site-directed mutagenesis method. This work was supported by a PhD CASE studentship from the Biotechnology and Biological Sciences Research Council and the National Institute for Biological Standards and Control to B.J.D., the Research Stimulation Fund (USF) of the Vrije Universiteit, Amsterdam, The Netherlands, and a grant of the Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO 901-14-206) to A.H.M.v.V.